Whale Boogers, Fat Seals, and 64,000 Turtles: A Tour of Drones in Marine Research

Marine biology has a data problem. The animals are enormous, the ocean is bigger, and most of your research subjects spend their lives underwater in places that are expensive, dangerous, or impossible to access. For decades, this meant marine research was slow, expensive, and often invasive - chasing whales in boats, shooting biopsy darts, deploying tagging crews, or waiting patiently for animals to wash up dead so you could finally get a good look at them.

Then drones showed up, and things got weird. In the best possible way.

Over the past decade, researchers have figured out that small aircraft with cameras can do things that boats and planes and patient biologists simply couldn't. The result is a wave of creative, sometimes bizarre, and genuinely groundbreaking research. We've been following this literature partly because we do marine mammal monitoring ourselves, and partly because it's just fascinating.

Here's a tour of some highlights.

Collecting Whale Snot (Yes, Really)

When a whale exhales, it blows more than just air. That spout contains mucus, lung cells, bacteria, hormones, and DNA - a biological treasure trove that researchers call "exhaled breath condensate" and everyone else calls whale snot.

For years, the only way to collect it was to lean over the side of a boat with a long pole and try to catch the blow before the whale dove. This worked about as well as you'd expect.

Enter SnotBot, a project from Ocean Alliance that started in 2015. The concept is elegant: mount petri dishes on top of a drone, fly it through the whale's blow as it surfaces, and collect the snot without the whale ever knowing you're there. The drone approaches from behind as the whale moves forward, the snot arcs up and gets pulled down onto the collection dishes, and the aircraft returns to the research vessel with samples intact.

What can you learn from whale boogers? Quite a lot, it turns out. DNA for population genetics and individual identification. Stress hormones that reveal how whales are responding to environmental pressures. Pregnancy hormones that indicate reproductive status. Microbiome samples that show what's living in their respiratory systems. All without chasing the animal, shooting it with anything, or stressing it out.

Ocean Alliance has now collected over 500 samples from five species across multiple continents, and the technique has spread to research groups worldwide. The IEEE Spectrum article on the project quotes one researcher's description of the old method: "Imagine if everything your doctor knew about your health came from chasing you around the room with a large needle while blowing an air horn."

SnotBot is the air-horn-free alternative.

The technique has made its way to Canadian waters as well. Research teams have been collecting blow samples from humpback whales around Vancouver Island under DFO permits, contributing to studies on whale microbiomes and respiratory health. Closer to home for us, Aeria's own Geoff Mullins built a custom SnotBot-style collection system forhis role with Fisheries and Oceans Canada here in BC - one of several examples of how the technology has been adapted for local research programs. A 2017 study published in mSystems compared blow microbiomes from humpbacks off Cape Cod and Vancouver Island, finding a shared core microbiome across both populations - the kind of finding that's only possible when researchers can collect samples non-invasively from whales that don't know they're being studied.

Putting Whales on a Scale (Without the Scale)

Here's a problem: you want to know if a whale is healthy. A fat whale is generally a healthy whale - it's been finding enough food to build energy reserves. A skinny whale is a whale in trouble. But how do you weigh an animal that can mass 40 tons and never holds still?

You don't. Instead, you measure it from above.

Drone-based photogrammetry has revolutionized whale health assessment. Researchers fly directly over surfacing whales, capture images at known altitudes, and use the altitude data to calculate actual body measurements. Length. Width at multiple points along the body. Surface area. Estimated volume.

From these measurements, they calculate something called Body Area Index (BAI) or Body Condition Index - essentially a whale BMI. The math is more complicated than human BMI because whales aren't uniform cylinders, but the principle is similar: compare an individual's measurements to what's normal for their length, and you get a sense of whether they're thriving or struggling.

This technique has produced genuinely important findings. Researchers at Oregon State's GEMM Lab have been tracking gray whale body condition for years, documenting how individuals gain and lose weight across seasons and years, and correlating those changes with environmental conditions, reproductive status, and prey availability. When gray whales experienced an Unusual Mortality Event from 2019–2023 with elevated strandings and declining birth rates, drone photogrammetry provided data showing the population's declining body condition - evidence that the whales weren't finding enough food.

Similarly, researchers studying killer whales have developed the "eye-patch ratio"—the angle between the white patches on a killer whale's head - as a proxy for body condition. Fatter whales have different proportions than skinnier whales, and these differences show up in overhead photographs. By tracking the same individuals over time, researchers can detect early signs of nutritional stress before it becomes a crisis.

The latest development: combining photogrammetry with data from biologging tags that measure tissue density. A 2025 paper in Ecology and Evolution demonstrated a method for estimating actual lipid mass in sperm whales - not just a proxy for body condition, but a direct measurement of fat reserves. The technique estimated lipid mass to within about 18% accuracy. For an animal you literally cannot weigh, that's remarkable.

How Low Can You Go? (The Seal Disturbance Question)

Drones are supposed to be less disturbing to wildlife than boats or planes. But how much less? And at what altitude do they start causing problems?

A 2024 study in Frontiers in Marine Science addressed this directly, examining how harbour seals and grey seals responded to drones at different altitudes. The researchers flew commercial drones (DJI Phantom 4 and Mavic series) over hauled-out seals in the Wadden Sea, documenting behavioural responses at various heights and approach angles.

The findings were nuanced. At 100+ meters, seals showed minimal response - the drone was essentially invisible to them. Below 50 meters, some individuals showed alert behaviours. Below 30 meters, the probability of flushing (seals leaving the haul-out) increased significantly. Approach angle mattered too: vertical descents were more disturbing than horizontal approaches at the same altitude.

The practical implication: there's a sweet spot where drones can collect useful data without disturbing the animals. For seals, that appears to be above 40–50 meters for most purposes. For cetaceans, studies suggest 30+ meters is generally safe. Different species have different thresholds, and environmental factors matter - wind affects sound propagation, and animals may be more sensitive during certain activities.

This research matters because it allows regulators and researchers to set evidence-based guidelines. Flying too high means poor data quality. Flying too low means disturbing the animals you're trying to study. The optimal altitude is somewhere in between, and it varies by species, context, and research objective.

Counting 64,000 Turtles

Raine Island, off the coast of Australia, hosts the largest green sea turtle rookery in the world. During peak nesting season, the beaches are so crowded with turtles that researchers struggled to count them using traditional methods - walking the beach at night, painting shells with stripes to mark individuals, then trying to count marked and unmarked turtles in the water the next day.

A 2020 study published in PLOS ONE compared this traditional approach to drone-based surveys. The results weren't close. Drones were faster, more accurate, and didn't require researchers to wade through crowds of nesting turtles in the dark. The drone footage captured more than 64,000 turtles in the waters around the island - a number that would have been essentially impossible to document any other way.

Beyond raw counting, drones have enabled entirely new types of sea turtle research. Thermal cameras can detect nesting turtles at night without disturbing them with lights. Photogrammetry can map beach profiles to understand how sea level rise will affect nesting habitat. Multispectral imagery can identify vegetation types that influence nest temperature and hatchling sex ratios (sea turtle sex is determined by incubation temperature).

A 2022 study from Greece used drones to reveal something that beach-based surveys couldn't: the dynamic positioning of breeding aggregations offshore. Turtles weren't randomly distributed in the water - their locations shifted with wind direction as females sought warmer nearshore waters to mature their eggs. Males and females showed different orientation patterns at different times of the season. None of this was visible from the beach.

The AI Layer

The latest development in drone-based marine research isn't about the drones at all - it's about what happens to the footage afterward.

Researchers at Oregon State have developed open-source AI tools called DeteX and XtraX that automate whale detection and measurement. DeteX processes drone video and automatically identifies frames containing whales. XtraX takes those frames and extracts body length and body condition measurements. What used to take hours of manual video review can now happen in minutes.

SharkEye, a project at UC Santa Barbara, is doing something similar for great white sharks - training computer vision models to automatically detect sharks in drone footage and eventually generate real-time alerts for beach safety.

The implication is that drone surveys can scale in ways that weren't previously possible. A research team can fly dozens of survey hours, let AI process the footage, and extract measurements from hundreds of individuals - generating datasets that would have taken years to compile using traditional methods.

What This Means

We do marine mammal monitoring ourselves - narwhal behaviour studies in the Arctic, pinniped surveys for salmon research, that kind of thing. We've watched this research space evolve, and what strikes us is how much the technology has democratized marine science.

A decade ago, studying whale health required research vessels, specialized equipment, and budgets measured in hundreds of thousands of dollars. Today, a team with a commercial drone, some training, and appropriate permits can collect data that contributes to global datasets on whale body condition, population dynamics, and behavioural ecology.

That doesn't mean it's easy - the research described here represents years of methodological development, careful validation, and hard-won operational expertise. Flying drones over marine mammals requires permits, protocols, and a genuine understanding of animal behaviour. The regulatory and ethical frameworks are still evolving.

But the trajectory is clear. Drones are giving researchers views of marine life that simply weren't available before. Whale snot. Turtle aggregations. Seal behaviour. Shark movements. All captured quietly, from above, without the disturbance of boats or the danger of close approaches.

The ocean is still vast and the animals are still elusive. But we're seeing more of it than ever before, one flight at a time.

---

Aeria Solutions conducts marine mammal monitoring programs across Canada, including BVLOS narwhal behaviour studies in the Arctic and pinniped surveys for salmon conservation research. We follow the marine research literature because it informs our own work - and because whale snot is genuinely interesting.